> ## Documentation Index
> Fetch the complete documentation index at: https://docs.coderabbit.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Test user journeys with CodeRabbit Verify

> Build a library of user journeys for a repository, then test them on a branch on demand or on a schedule.

GitHub Cloud (Verify is available on GitHub Cloud. GitHub Enterprise Server is not supported.) GitLab.com (Verify is available on GitLab.com. GitLab self-managed is not supported.) Bitbucket Cloud (Verify is available on Bitbucket Cloud. Bitbucket Data Center is not supported.) | Cloud Only (This feature is available on CodeRabbit Cloud only. It is not available on self-hosted CodeRabbit deployments.) Open Beta (Verify is in beta and free during the beta.)

A user journey is a saved test of a flow your users depend on, such as signing in or checking out. Each repository keeps its own library of journeys. You run them against a branch, on demand or on a schedule, to check that those flows still work. Journey runs test a branch and don't post on pull requests; pull request runs use scenarios that Verify writes from the change instead. See [Test pull requests](https://docs.coderabbit.ai/verify/pull-requests.md).

To find a repository's journeys, select **Verify** in the sidebar, choose the repository, and select **Journey Testing** > **User journeys**. Creating, editing, and running journeys requires the Coding Agent write permission. See [Permissions](https://docs.coderabbit.ai/verify/access-and-billing.md#permissions).

## What a journey contains

| Field | What it holds | Limit |
| - | - | - |
| **Title** | A short name for the flow. | 1 to 300 characters |
| **User / role** | Who completes the journey, for example a workspace admin. | Up to 4,096 characters |
| **Goal** | The outcome this person should achieve. | Up to 10,000 characters |
| **Steps** | The actions in order. Keep each action in its own step. | 1 to 50 steps, each up to 4,096 characters |
| **Assertions** | Optional conditions the tester must verify true for the journey to pass. | Up to 50, each up to 4,096 characters |
| **Tags** | Optional labels for selecting journeys in runs and schedules. | Up to 50, each 1 to 100 characters |

A repository's library holds up to 50 journeys.

## Create journeys

You can create journeys in three ways:

* **Map with an Agent**: on an empty library, select **Map with an Agent**. An agent inspects the repository, without running its code or changing it, and proposes journeys for the branch. Open its conversation with **View Agent**, review the proposal, and tell the agent which journeys to add.
* **Add user journey**: select **Add user journey** and fill in the drawer. **Title**, **User / role**, **Goal**, and at least one step are required. Select **Add to library** to save it.
* **Duplicate journey**: in a journey's row actions, select **Duplicate journey** to start from a copy.

To have an agent revise the whole library later, select **Edit with Agent** on the **User journeys** page. To revise one journey, open it and select **Manage journey** > **Edit with agent**.

## Edit journeys

Open a journey and select **Edit**, or select **Edit journey** in its row actions. Change it, and select **Save user journey**. When an edit changes more than one thing, the **Review journey changes** dialog summarizes the changes before you save.

Each save creates a new revision of the journey. Results from earlier runs keep the definition they ran with, and a result from an older revision is marked **Outdated**.

To delete a journey, select **Manage journey** > **Delete user journey**, or select several journeys and choose **Delete selected**. Deleting a journey also deletes its linked bugs and blockers. Its execution history remains available.

## Run journeys on demand

**1. Open the run dialog**

On the repository's **User journeys** page, select **Run tests** on the toolbar. It appears while no journeys are selected. To run specific journeys, select them and choose **Run selected**, or open a journey and select **Run**. The **Run verifier** dialog opens.

**2. Choose the branch**

In **Run branch**, choose the branch to test. The environment comes from the repository's settings; change it there if you need another one.

**3. Choose the journeys**

Select **Tags** to run only the journeys that have all the selected tags, or pick journeys under **Choose journeys for this run**. **Include all** and **Clear all** select or clear the list. If you filtered the journey table, **Run scope** chooses between **Filtered journeys** and **All active journeys**. A run can include up to 50 journeys.

**4. Start the run**

Select the **Run** button, which shows how many journeys will run.

Only one run at a time can be active for a branch and environment. While one is queued or running, the **Run verifier** dialog says "A verifier is already queued or running for this branch and environment. Wait for it to finish before starting another run." and you can't start another.

Each journey in a run gets its own sandbox, restored from the prepared environment, and the journeys run in parallel. They don't share app state, so a journey can't depend on another journey's steps or data.

While the run is active, its page lists the run's journeys, each with a status. Select a journey to follow its tester; a banner shows its state, such as **Preparing test**, **Testing journey**, or **Test finished**. Journey runs run on their own, so you can't send messages to the tester. To stop the run, select **Stop run**.

Results appear on the repository's **Journey Testing** > **Run history** page, in each journey's **Latest run**, and on the Activity page. See [Results and bugs](https://docs.coderabbit.ai/verify/results.md).

## Schedule runs

A schedule runs a set of journeys on a branch regularly. On the repository's page, select **Journey Testing** > **Schedules**, then select **Create schedule** (or **Create first schedule** if there are none yet).

| Setting | What it does |
| - | - |
| **Name** | Optional, up to 200 characters. |
| **Journeys to run** | **Specific journeys**, up to 50, or **Matching tags**. With tags, the matching journeys are worked out again at each run, so new journeys with those tags are included. |
| **Branch** | The branch to test. |
| **Environment** | A ready environment for the repository. |
| **Frequency** | **Every day**, **Every week**, **Every month**, or **Every N minutes**. |
| **Time**, **Days of week**, **Day of month** | When calendar schedules run. With **Day of month**, months without that day are skipped. |
| **Interval in minutes** | For **Every N minutes**: a whole number from 15 to 525,600. |
| **Timezone** | An IANA timezone, such as `Europe/Berlin`, that the schedule's times use. |
| **Auto create issues** | Optionally create Jira or Linear issues for new findings. See [Automatic tickets from schedules](https://docs.coderabbit.ai/verify/results.md#automatic-tickets-from-schedules). |

How schedules behave:

* A scheduled run runs as the person who created the schedule. Their access is checked again at each occurrence.
* Each occurrence tests the head of the branch when it starts. The journey's run history shows the commit it tested.
* If another Verify run is active on the same branch and environment, that occurrence is skipped, not queued.
* Use a schedule's row actions to **Edit**, **Pause** or **Resume**, or **Delete** it. Pausing and deleting stop future occurrences; runs already accepted can finish.
* Scheduled runs appear in the repository's run history like on-demand runs.
* Verify doesn't notify anyone when a scheduled run fails, is blocked, or is skipped. Check the schedule's [Occurrence history](#occurrence-history), or turn on **Auto create issues** to get Jira or Linear issues for new bugs, or for new bugs and blockers with **Bugs & blockers**. An issue is created only for a finding first seen in that run. Failed and skipped occurrences don't create issues.

### Occurrence history

Select **Occurrence history** in a schedule's row actions to see each occurrence and why one was skipped or failed:

* "No journeys matched the saved tags."
* "A selected journey or repository is unavailable."
* "Another Verify run was active on this branch and environment."
* "The saved environment is unavailable or needs preparation."
* "The saved branch is unavailable."
* "The schedule creator no longer has the required access."
* "Verify execution could not pass the billing checks."
* "The journey library or execution configuration changed before admission."

## What's next

> [card] [Results and bugs](https://docs.coderabbit.ai/verify/results.md)
>
> Read outcomes and evidence, and turn bugs into tickets or coding tasks.

> [card] [Set up Verify](https://docs.coderabbit.ai/verify/setup.md)
>
> Change a repository's branch, environment, and test instructions.

> [card] [Troubleshooting](https://docs.coderabbit.ai/verify/troubleshooting.md)
>
> Fix runs that don't start or end without a result.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.